| { | |
| "best_metric": 0.14893724024295807, | |
| "best_model_checkpoint": "./lora-out/checkpoint-108", | |
| "epoch": 7.646017699115045, | |
| "eval_steps": 4, | |
| "global_step": 108, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 0.3006412982940674, | |
| "eval_runtime": 54.7333, | |
| "eval_samples_per_second": 0.914, | |
| "eval_steps_per_second": 0.238, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 0.300335168838501, | |
| "eval_runtime": 54.8113, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3024, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": 0.2993900179862976, | |
| "eval_runtime": 54.782, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "eval_loss": 0.29816487431526184, | |
| "eval_runtime": 54.8049, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3035, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_loss": 0.29595255851745605, | |
| "eval_runtime": 54.7885, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 0.2939557135105133, | |
| "eval_runtime": 54.7999, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_loss": 0.29013773798942566, | |
| "eval_runtime": 54.7805, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2959, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "eval_loss": 0.28251081705093384, | |
| "eval_runtime": 54.7706, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 0.2771329879760742, | |
| "eval_runtime": 54.7818, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 4e-05, | |
| "loss": 0.284, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "eval_loss": 0.27146488428115845, | |
| "eval_runtime": 54.803, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_loss": 0.26464152336120605, | |
| "eval_runtime": 54.8467, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "eval_loss": 0.25653430819511414, | |
| "eval_runtime": 54.8327, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 5e-05, | |
| "loss": 0.263, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "eval_loss": 0.24627122282981873, | |
| "eval_runtime": 54.813, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "eval_loss": 0.23474617302417755, | |
| "eval_runtime": 54.7901, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 6e-05, | |
| "loss": 0.241, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "eval_loss": 0.2220366895198822, | |
| "eval_runtime": 54.7983, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "eval_loss": 0.20926769077777863, | |
| "eval_runtime": 54.7403, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "eval_loss": 0.19629451632499695, | |
| "eval_runtime": 54.7525, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 7e-05, | |
| "loss": 0.2101, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_loss": 0.18524658679962158, | |
| "eval_runtime": 54.7303, | |
| "eval_samples_per_second": 0.914, | |
| "eval_steps_per_second": 0.238, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "eval_loss": 0.17731742560863495, | |
| "eval_runtime": 54.7552, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 8e-05, | |
| "loss": 0.1788, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "eval_loss": 0.16993452608585358, | |
| "eval_runtime": 54.729, | |
| "eval_samples_per_second": 0.914, | |
| "eval_steps_per_second": 0.238, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.95, | |
| "eval_loss": 0.164781853556633, | |
| "eval_runtime": 54.741, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "eval_loss": 0.16103117167949677, | |
| "eval_runtime": 54.7837, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 9e-05, | |
| "loss": 0.1615, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "eval_loss": 0.15781742334365845, | |
| "eval_runtime": 54.7138, | |
| "eval_samples_per_second": 0.914, | |
| "eval_steps_per_second": 0.238, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "eval_loss": 0.15516981482505798, | |
| "eval_runtime": 54.7516, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1533, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "eval_loss": 0.15261690318584442, | |
| "eval_runtime": 54.6891, | |
| "eval_samples_per_second": 0.914, | |
| "eval_steps_per_second": 0.238, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "eval_loss": 0.15066812932491302, | |
| "eval_runtime": 54.6884, | |
| "eval_samples_per_second": 0.914, | |
| "eval_steps_per_second": 0.238, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "eval_loss": 0.14893724024295807, | |
| "eval_runtime": 54.6275, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 108 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 210, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 12, | |
| "total_flos": 3.553655402645422e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |